This multicenter observational study aims to develop and validate an artificial intelligence foundation model for frozen-section pathology. The study includes a retrospective phase and a prospective validation phase. Retrospective frozen-section pathology data will be used for model development, internal validation, and external validation. A prospective multicenter cohort of patients undergoing intraoperative frozen-section examination will then be enrolled to evaluate the model in a real-world clinical setting. The model will analyze digitized frozen-section whole-slide images and will be evaluated for prespecified frozen-section pathology diagnostic tasks across multiple organ systems. Its performance will be assessed using pathological reference standards. The primary outcome is the area under the receiver operating characteristic curve. Secondary outcomes include accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. This study is observational and will not require research-mandated changes to routine clinical care.
Frozen-section pathology is an essential component of intraoperative consultation and provides rapid diagnostic information that may support surgical decision-making. However, frozen-section images differ substantially from formalin-fixed, paraffin-embedded tissue sections because of freezing artifacts, variable section quality, staining variation, and differences in tissue morphology. Most currently available pathology foundation models have been developed primarily using formalin-fixed, paraffin-embedded pathology images and may have limited generalizability to frozen-section pathology. This multicenter, observational study is designed to develop and validate an artificial intelligence foundation model for frozen-section pathology. The study consists of a retrospective phase and a prospective validation phase. In the retrospective phase, patients with frozen-section pathology data collected from January 2019 to April 2026 will be included for model development, internal validation, and external validation. The retrospective development and internal validation datasets will include patients from Sun Yat-sen Memorial Hospital, Sun Yat-sen University Cancer Center, and the Third Affiliated Hospital of Sun Yat-sen University. Retrospective external validation will include patients from collaborating hospitals, including Shantou Central Hospital, the First Affiliated Hospital of Shantou University Medical College, and the Fifth Affiliated Hospital of Sun Yat-sen University. In the prospective phase, patients undergoing intraoperative frozen-section examination will be enrolled from June 2026 to October 2026, including patients from Sun Yat-sen Memorial Hospital and patients from the Fifth Affiliated Hospital of Sun Yat-sen University. The study population will include patients with benign or malignant diseases involving the lung, thyroid, stomach, colorectum, liver, bladder, prostate, kidney, uterus, and ovary. For each eligible patient, clinical information, frozen-section whole-slide images, intraoperative pathology reports, and corresponding pathological data will be collected. Whole-slide images will undergo tissue detection, region-of-interest segmentation, and patch-level image processing. The model will be pretrained using self-supervised learning and subsequently fine-tuned for prespecified downstream frozen-section pathology tasks using multiple-instance learning or Transformer-based aggregation methods. Model interpretability will be explored using gradient-weighted class activation mapping. The primary outcome will be the discriminative performance of the artificial intelligence foundation model, assessed using the area under the receiver operating characteristic curve. Secondary outcomes will include accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and robustness across participating centers, organ systems, and image-quality conditions.
Study Type
OBSERVATIONAL
Enrollment
33,000
Sun Yat-sen Memorial Hospital of Sun Yat-sen University, Guangzhou, Guangdong
Guangzhou, China
Area Under the Receiver Operating Characteristic Curve
The area under the receiver operating characteristic curve (AUROC) will be calculated to evaluate the discriminative performance of the artificial intelligence foundation model for each prespecified frozen-section pathology diagnostic or prediction task. The reference standard will be the corresponding pathological diagnosis specified in the study protocol and statistical analysis plan. Higher AUROC values indicate better discriminative performance.
Time frame: For each enrolled patient, the diagnosis results of AI model will be obtained in several days after intraoperative pathology completion, and the AUROC of the AI model will be evaluated through study completion, an average of 3 year.
Diagnostic Accuracy
Diagnostic accuracy will be calculated as the proportion of correctly classified cases among all evaluable cases for each prespecified frozen-section pathology task at the prespecified operating threshold.
Time frame: For each enrolled patient, the diagnosis results of AI model will be obtained in several days after intraoperative pathology completion, and the accuracy of the AI model will be evaluated through study completion, an average of 3 years.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.